检索结果-内蒙古大学图书馆

Dermatological Diagnostics: A Unified Deep Learning Framework for Skin Lesion and Cancer Classification 1

学校读者我要写书评

暂无评论

Dermatological Diagnostics: A Unified Deep Learning Framewor...

1st IEEE International Conference on Advances in Computing, Communication and Networking, ICAC2N 2024

作者： Kumar, Vinod Jain, Isha Dongla, Durga Sritha Rai, Shivangi Rastogi, Vani Kaushik, Eva Kaushik, Rohit George Washington University School of Engineering and Applied Science WashingtonDC United States Dr. Akhilesh Das Gupta Institute of Technology and Management Computer Science and Engineering Department New Delhi India Gandhi Institute of Technology and Management Computer Science and Engineering Department Andhra Pradesh Visakhapatnam India Data Scientist DXC Technology Noida India Data Analytics Program University of Illinois CS and Mathematics Department Springfield United States

ISBN: (纸本)9798350356816

Skin pathologies encompass a spectrum of conditions, with malignancies such as melanoma representing a critical diagnostic urgency. This investigation delineates the deployment of Convolutional Neural Networks (CNNs) for the classification of dermatological anomalies, benchmarking CNN diagnostic fidelity against dermatological expert evaluations. The study underscores the efficacy of CNN classifiers in expediting the diagnostic workflow for various cutaneous disorders. Advocating for an automated diagnostic framework, the research introduces a CNN-based system aimed at reducing human diagnostic load, accelerating diagnostic timelines, and enhancing survival outcomes. Utilizing advanced image processing algorithms and deep learning architectures, the research presents an automated classification system for skin pathologies, addressing both benign and malignant presentations. The classification matrix includes nine dermatological conditions: actinic keratosis, basal cell carcinoma, benign keratosis, dermatofibroma, melanoma, nevus, seborrheic keratosis, squamous cell carcinoma, and vascular lesions. The objective is to engineer a CNN model with robust diagnostic performance across a diverse lesion dataset, ensuring accurate identification and categorization of dermatological conditions. © 2024 IEEE.

关键词： Convolutional neural networks

Precision spectral line-resolution to classify greenhouse gas species in broadband mid-infrared sensing

学校读者我要写书评

暂无评论

Precision spectral line-resolution to classify greenhouse ga...

Conference on Lasers and Electro-Optics (CLEO)

作者： Al- Alexis Zayna Juracka Zakaria Juracka Amir Khan Division of Physics Engineering Mathematics and Computer Science Optical Science Center for Applied Research Delaware State University Dover DE

We present a novel methodology to classify greenhouse gas species by investigating the structural complexity of broadband overlapping molecular lines in the spectral region of 7.0 μm to 8.0 μm. The structural inform... 详细信息

ISBN: (纸本)9781957171258

关键词： Absorption spectroscopy Infrared lasers Laser beams Laser scanning Light intensity Quantum cascade semiconductor lasers

Multilevel Particle Filters for a Class of Partially Observed Piecewise Deterministic Markov Processes

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Jasra, Ajay Kamatani, Kengo Maama, Mohamed Applied Mathematics and Computational Science Program Computer Electrical and Mathematical Sciences and Engineering Division King Abdullah University of Science and Technology Thuwal23955-6900 Saudi Arabia Institute of Statistical Mathematics Tokyo190-0014 Japan

In this paper we consider the filtering of a class of partially observed piecewise deterministic Markov processes (PDMPs). In particular, we assume that an ordinary differential equation (ODE) drives the deterministic element and can only be solved numerically via a time discretization. We develop, based upon the approach in [20], a new particle and multilevel particle filter (MLPF) in order to approximate the filter associated to the discretized ODE. We provide a bound on the mean square error associated to the MLPF which provides guidance on setting the simulation parameter of that algorithm and implies that significant computational gains can be obtained versus using a particle filter. Our theoretical claims are confirmed in several numerical examples. Copyright © 2023, The Authors. All rights reserved.

关键词： Ordinary differential equations

A Pseudo-Spectral Scheme for Systems of Two-Point Boundary Value Problems with Left and Right Sided Fractional Derivatives and Related Integral Equations

学校读者我要写书评

暂无评论

computer Modeling in Engineering & sciences 2021年第7期128卷 21-41页

作者： I.G.Ameen N.A.Elkot M.A.Zaky A.S.Hendy E.H.Doha Department of Mathematics Faculty of ScienceAl-Azhar UniversityCairoEgypt Department of Mathematics Faculty of ScienceCairo UniversityGiza12613Egypt Department of Applied Mathematics Physics DivisionNational Research CentreDokkiCairo12622Egypt Department of Computational Mathematics and Computer Science Institute of Natural Sciences and MathematicsUral Federal UniversityYekaterinburg620002Russia Department of Mathematics Faculty of ScienceBenha UniversityBenha13511Egypt

We target here to solve numerically a class of nonlinear fractional two-point boundary value problems involving left-and right-sided fractional *** main ingredient of the proposed method is to recast the problem into an equivalent system of weakly singular integral ***,a Legendre-based spectral collocation method is developed for solving the transformed ***,we can make good use of the advantages of the Gauss quadrature *** present the construction and analysis of the collocation *** results can be indirectly applied to solve fractional optimal control problems by considering the corresponding Euler–Lagrange *** numerical examples are given to confirm the convergence analysis and robustness of the scheme.

关键词： Spectral collocation method weakly singular integral equations two-point boundary value problems convergence analysis

Unbiased Parameter Estimation for Bayesian Inverse Problems

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Chada, Neil K. Jasra, Ajay Maama, Mohamed Tempone, Raul Department of Mathematics City University of Hong Kong China School of Data Science The Chinese University of Hong Kong Shenzhen CN Shenzhen China Applied Mathematics and Computational Science Program Computer Electrical and Mathematical Sciences and Engineering Division King Abdullah University of Science and Technology Thuwal23955-6900 Saudi Arabia

In this paper we consider the estimation of unknown parameters in Bayesian inverse problems. In most cases of practical interest, there are several barriers to performing such estimation, This includes a numerical approximation of a solution of a differential equation and, even if exact solutions are available, an analytical intractability of the marginal likelihood and its associated gradient, which is used for parameter estimation. The focus of this article is to deliver unbiased estimates of the unknown parameters, that is, stochastic estimators that, in expectation, are equal to the maximizer of the marginal likelihood, and possess no numerical approximation error. Based upon the ideas of [4] we develop a new approach for unbiased parameter estimation for Bayesian inverse problems. We prove unbiasedness and establish numerically that the associated estimation procedure is faster than the current state-of-the-art methodology for this problem. We demonstrate the performance of our methodology on a range of problems which include a PDE and ODE. Copyright © 2025, The Authors. All rights reserved.

关键词： Markov processes

Prognostic Kalman Filter Based Bayesian Learning Model for Data Accuracy Prediction

学校读者我要写书评

暂无评论

computers, Materials & Continua 2022年第7期72卷 243-259页

作者： S.Karthik Robin Singh Bhadoria Jeong Gon Lee Arun Kumar Sivaraman Sovan Samanta A.Balasundaram Brijesh Kumar Chaurasia S.Ashokkumar Department of ECE College of Engineering and TechnologySRM Institute of Science and TechnologyVadapalaniChennai600026India Department of CSE Birla Institute of Applied Sciences(BIAS)BhimtalUttarakhand263136India Division of Applied Mathematics Wonkwang University460Iksan-DaeroIksan-SiJeonbuk54538Korea School of Computer Science and Engineering Vellore Institute of Technology(VIT)Chennai600127India Department of Mathematics Tamralipta MahavidyalayaWest Bengal721636India School of Computer Science and Engineering Center for Cyber Physical SystemsVellore Institute of Technology(VIT)Chennai600127India Indian Institute of Information Technology(IIIT) LucknowUttar Pradesh226002India Department of Computer Science and Engineering Saveetha School of EngineeringSIMATSChennai602105India

Data is always a crucial issue of concern especially during its prediction and computation in digital *** paper exactly helps in providing efficient learning mechanism for accurate predictability and reducing redundant data *** also discusses the Bayesian analysis that finds the conditional probability of at least two parametric based predictions for the *** paper presents a method for improving the performance of Bayesian classification using the combination of Kalman Filter and *** method is applied on a small dataset just for establishing the fact that the proposed algorithm can reduce the time for computing the clusters from *** proposed Bayesian learning probabilistic model is used to check the statistical noise and other inaccuracies using unknown *** scenario is being implemented using efficient machine learning algorithm to perpetuate the Bayesian probabilistic *** also demonstrates the generative function forKalman-filer based prediction model and its *** paper implements the algorithm using open source platform of Python and efficiently integrates all different modules to piece of code via Common Platform Enumeration(CPE)for Python.

关键词： Bayesian learning model kalman filter machine learning data accuracy prediction

An Improved Unbiased Particle Filter

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Jasra, Ajay Maama, Mohamed Ombao, Hernando Applied Mathematics and Computational Science Program Computer Electrical and Mathematical Sciences and Engineering Division King Abdullah University of Science and Technology Thuwal23955-6900 Saudi Arabia Statistics Program Computer Electrical and Mathematical Sciences and Engineering Division King Abdullah University of Science and Technology Thuwal23955-6900 Saudi Arabia

In this paper we consider the filtering of partially observed multi-dimensional diffusion processes that are observed regularly at discrete times. We assume that, for numerical reasons, one has to time-discretize the diffusion process which typically leads to filtering that is subject to discretization bias. The approach in [16] establishes that when only having access to the time discretized diffusion it is possible to remove the discretization bias with an estimator of finite variance. We improve on the method in [16] by introducing a modified estimator based on the recent work of [17]. We show that this new estimator is unbiased and has finite variance. Moreover, we conjecture and verify in numerical simulations that substantial gains are obtained. That is, for a given mean square error (MSE) and a particular class of multi-dimensional diffusion, the cost to achieve the said MSE falls. Copyright © 2023, The Authors. All rights reserved.

关键词： Mean square error

Applying Layer-Wise Relevance Propagation on U-Net Architectures 27th

学校读者我要写书评

暂无评论

Applying Layer-Wise Relevance Propagation on U-Net Architect...

27th International Conference on Pattern Recognition, ICPR 2024

作者： Weinberger, Patrick Fröhler, Bernhard Heim, Anja Gall, Alexander Bodenhofer, Ulrich Senck, Sascha Research Group Computed Tomography University of Applied Sciences Upper Austria Wels Austria School of Informatics Communication and Media University of Applied Sciences Upper Austria Wels Austria Johannes Kepler University Linz Linz Austria Faculty of Computer Science and Mathematics University of Passau Passau Germany Division Development Center X-ray Technology Fraunhofer Institute for Integrated Circuits IIS Erlangen Germany

ISBN: (纸本)9783031781971

For safety critical applications, it is still a challenge to use AI and fulfill all regulatory requirements. Medicine/healthcare and transportation are two fields where regulatory requirements are of fundamental importance. A wrong decision can lead to serious hazards or even deaths. In these fields, semantic segmentation is often utilized to extract features. Especially U-Net architectures are used. This paper shows how to apply layer-wise relevance propagation (LRP) to a trained U-Net architecture. We achieve an efficient explanation of a segmentation by back-propagating the whole resulting image. To tackle the non-linear results of the LRP, we introduce a threshold mechanism in combination with a logarithmic transfer function to preprocess the data for visualization. We demonstrate our method on three use cases: the segmentation of a fiber-reinforced polymer in the field of non-destructive testing, the segmentation of pedestrians in an automotive application, and a lung segmentation example from the medical domain. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Semantic Segmentation

Developing, Analyzing, and Evaluating Vehicular Lane Keeping Algorithms Under Dynamic Lighting and Weather Conditions Using Electric Vehicles

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Khalfin, Michael Volgren, Jack Jones, Matthew LeGoullon, Luke Siegel, Joshua Chung, Chan-Jin Department of Computational Applied Mathematics Rice University United States Department of Computer Science and Engineering Pennsylvania State University United States Department of Mathematics and Computer Science Willamette University United States Division of Computer Science and Engineering Louisiana State University United States Department of Computer Science and Engineering Michigan State University United States Department of Mathematics and Computer Science Lawrence Technological University United States

Self-driving vehicles have the potential to reduce accidents and fatalities on the road. Many production vehicles already come equipped with basic self-driving capabilities, but have trouble following lanes in adverse lighting and weather conditions. Therefore, we develop, analyze, and evaluate two vehicular lane-keeping algorithms under dynamic weather conditions using a combined deep learning- and hand-crafted approach and an end-to-end deep learning approach. We use image segmentation- and linear-regression based deep learning to drive the vehicle toward the center of the lane, measuring the amount of laps completed, average speed, and average steering error per lap. Our hybrid model completes more laps than our end-to-end deep learning model. In the future, we are interested in combining our algorithms to form one cohesive approach to lane-following. © 2024, CC BY.

关键词： computer vision